Electromagnetic transducers such as heads for disk or tape drives commonly include Permalloy (approximately Ni0.81 Fe0.19), which is formed in thin layers to create magnetic features. Permalloy is known to be magnetically “soft,” that is, to have high permeability and low coercivity, allowing structures made of Permalloy to act like good conductors of magnetic flux. Disks having a media layer that stores magnetic bits in a direction substantially perpendicular to the media surface, sometimes termed “perpendicular recording,” have been proposed to have a soft magnetic underlayer of permalloy or the like.
For example, an inductive head may have conductive coils that induce magnetic flux in an adjacent Permalloy core, that flux employed to magnetize a portion or bit of an adjacent media. That same inductive head may read signals from the media by bringing the core near the magnetized media portion so that the flux from the media portion induces a flux in the core, the changing flux in the core inducing an electric current in the coils. Alternatively, instead of inductively sensing media fields, magnetoresistive (MR) sensors or merged heads that include MR or giant magnetoresistive (GMR) sensors may use thinner layers of Permalloy to read signals, by sensing a change in electrical resistance of the sensor that is caused by the magnetic signal. For perpendicular recording, the soft magnetic underlayer of the disk as well as the soft magnetic core of the head may together form a magnetic circuit for flux that travels across the media layer to write or read information.
In order to store more information in smaller spaces, transducer elements have decreased in size for many years. One difficulty with this deceased size is that the amount of flux that needs to be transmitted may saturate elements such as magnetic pole layers, which becomes particularly troublesome when ends of the pole layers closest to the media, commonly termed pole tips, are saturated. Magnetic saturation in this case limits the amount of flux that is transmitted through the pole tips, limiting writing or reading of signals. Moreover, such saturation may blur that writing or reading, as the flux may be evenly dispersed over an entire pole tip instead of being focused in a corner that has relatively high flux density. For these reasons the use of high magnetic saturation materials (also known as high moment or high BS materials) in magnetic core elements has been known for many years.
For instance, iron is known to have a higher magnetic moment than nickel, so increasing the proportion of iron compared to nickel generally yields a higher moment alloy. Iron, however, is also more corrosive than nickel, which imposes a limit to the concentration of iron that is feasible for many applications. Also, it is difficult to achieve soft magnetic properties for primarily-iron NiFe compared to primarily-nickel NiFe. Anderson et al., in U.S. Pat. No. 4,589,042, teach the use of high moment Ni0.45Fe0.55 for pole tips. Anderson et al. do not use Ni0.45Fe0.55 throughout the core due to problems with permeability of that material, which Anderson et al. suggest is due to relatively high magnetostriction of Ni0.45Fe0.55.
As noted in U.S. Pat. No. 5,606,478 to Chen et al., the use of high moment materials has also been proposed for layers of magnetic cores located closest to a gap region separating the cores. Also noted by Chen et al. are some of the difficulties presented by these high moment materials, including challenges in forming desired elements and corrosion of the elements once formed. Chen et al. state that magnetostriction is another problem with Ni0.45Fe0.55, and teach the importance of constructing of heads having Permalloy material layers that counteract the effects of that magnetostriction. This balancing of positive and negative magnetostriction with plural NiFe alloys is also described in U.S. Pat. No. 5,874,010 to Tao et al.
Primarily iron FeCo alloys are known to have a very high saturation magnetization but also high magnetostriction that makes them unsuitable for many head applications. That is, mechanical stress during slider fabrication or use may perturb desirable magnetic domain patterns of the head. FIG. 7 shows a B/H loop 12 of a FeCoN layer that was formed by sputtering deposition at room temperature, the layer having a thickness of approximately 500 Å and having a composition of approximately Fe0.66Co0.28N0.06. The applied H-field is shown in oersted (Oe) across the horizontal axis while the magnetization of the layer is plotted in normalized units along the vertical axis, with unity defined as the saturation magnetization for a given material. The FeCoN layer has a saturation magnetization (Bs) of approximately 24.0 kilogauss and is magnetically isotropic, as shown by the single B/H loop 12. B/H loop 12 also indicates a relatively high coercivity of about 80 oersted, which may be unsuitable for applications requiring soft magnetic properties.
In an article entitled “Microstructures and Soft Magnetic Properties of High Saturation Magnetization Fe—Co—N alloy Thin Films,” Materials Research Society, Spring meeting, Section F, April 2000, N. X. Sun et al. report the formation of FeCoN films having high magnetic saturation but also high magnetostriction and moderate coercivity. Sun et al. also report the formation of a thin film structure in which FeCoN is grown on and capped by Permalloy, to create a sandwich structure having reduced coercivity but compressive stress. The magnetostriction of this sandwich structure, while somewhat less than that of the single film of FeCoN, may still be problematic for head applications. Such issues would be expected to grow with increased length of a magnetostrictive layer, so that disk layers that extend many times as far as head layers would appear to be poor candidates for magnetostrictive materials.